agentleFS
YiShu5YiShu5/claude-skills/skills/HTMLHero.skill/SKILL.md

HTMLHero.skill

Skill131 starsChanged 54 days agoMITView on GitHub

Build or extend a runnable, responsive Hero first screen for an HTML/React site or web presentation when the user provides a Hero brief, visual reference, media, or existing frontend. Use this for the Hero itself; use HTMLPPT.skill for a multi-slide deck.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use for a new or changed Hero first screen with layout, copy, media, or interaction.

    line 12
  2. Very specific:

    If the request is a multi-slide HTML presentation, use HTMLPPT.skill instead.

    line 13
  3. Distinctive:

    If the request is only visual ideation or copy, do not force an implementation workflow.

    line 14
  4. Distinctive:

    Read references/validation.md for the relevant checks.

    line 23

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

HTMLHero.skill

Turn a Hero brief or an existing first-screen implementation into a runnable, reviewable Hero. The first screen should establish a visual system that can be extended later.

Route

  • Use for a new or changed Hero first screen with layout, copy, media, or interaction.
  • If the request is a multi-slide HTML presentation, use HTMLPPT.skill instead.
  • If the request is only visual ideation or copy, do not force an implementation workflow.

Workflow

  1. Establish the finish line and scope. Inspect the target directory, recent AGENTS.md, package.json, lockfile, relevant entrypoints, and referenced assets. Do not read the whole repository by default.
  2. Resolve the smallest useful Hero contract: purpose and audience, required copy, visual references, media, interactions, target viewport, existing stack, and delivery target. Read references/hero-contract.md when inputs are incomplete or the work will be handed off.
  3. Apply constraints in this order: user request, existing design system and project conventions, then skill defaults. Make reversible assumptions and state them. Ask only when a missing decision materially changes the result or crosses a real permission boundary; a routine local implementation does not need a ritual confirmation.
  4. Implement in the existing stack and file structure. For a new project, choose the smallest viable starter for the requested output; do not force React, Tailwind, animation libraries, or new dependencies. Preserve unrelated dirty work and avoid overwriting files outside scope.
  5. Make the Hero usable without unavailable assets: provide a poster, gradient, local fallback, or explicit placeholder when appropriate. Keep CTA targets meaningful, use semantic HTML, and treat animation as optional and reduced-motion aware.
  6. Read references/validation.md for the relevant checks. Continue through build, preview, visual inspection, and fixes until the finish line is met or a concrete blocker remains.

Completion and delivery

A task is complete when the requested Hero is implemented, the available build/preview path has been exercised, the target viewport has been inspected, and material failures have been fixed or clearly reported. Do not stop after the first implementation merely to request review when the remaining work is safe and in scope.

Report the changed files, actual run/build commands and results, preview URL or entrypoint, visual/interaction checks, assumptions or fallbacks, and any concrete blocker. Only create a zip or publish when requested.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.